632

IEEE TRANSACTIONS ON SMART GRID, VOL. 5, NO. 2, MARCH 2014

Location-Based Forecasting of Vehicular Charging

Load on the Distribution System
Nima Ghiasnezhad Omran, Student Member, IEEE, and Shaahin Filizadeh, Senior Member, IEEE

AbstractThis paper presents a procedure for location-based

forecasting of the potential vehicular charging load at off-home
charging stations. A location-specific fuzzy decision making system
is proposed to characterize the charging behavior and determine
the probability of charging by means of a three dimensional input
obtained from a real-world driving dataset. The obtained charging
and parking probability figures are then used in prediction of the
local aggregated charging demand. The flexibility and usefulness
of the developed procedure is exemplified in the case studies of two
major shopping centers.
Index TermsFuzzy decision making, location-based prediction, off-home charging station, vehicular charging demand.

I. INTRODUCTION

EHICULAR charging load on the grid is expected to rise

as plug-in (hybrid) electric vehicles (PEVs) experience
rapid improvements, which will lead to their increased performance, lower price and hence further public acceptance [1][3].
In the context of the future smart grid, grid-connected vehicular
storage systems are envisioned not only to require energy, but
also to be potentially contributing to the supply of energy during
periods of high demand [3][5].
Estimation of the aggregated vehicular demand on a specific
power system helps in the planning of generation to keep pace
with the growth of the load as PEVs gain larger market shares
[5][7]. It must, however, be noted that vehicular charging may
lead to severe issues at the distribution level even at low market
penetration levels and prior to manifesting any major problems
on the generation or transmission level [5][9]. Thus the distributed nature of the charging load and its localized impact on
the assets of the grid must be evaluated in the planning of the
distribution system.
In order to investigate the intensity of these potential issues, a
reliable prediction of the profile of the charging demand on the
distribution network is essential. Conventional load forecasting
methods in power systems for different time horizons employ
various statistical and artificial intelligence techniques (e.g.,
Manuscript received March 12, 2013; revised June 21, 2013 and August 21,
2013; accepted September 17, 2013. Date of publication December 05, 2013;
date of current version February 14, 2014. Financial support for this work was
provided by the Natural Sciences and Engineering Research Council (NSERC)
of Canada. Paper no. TSG-00212-2013.
The authors are with the Department of Electrical and Computer Engineering, University of Manitoba, Winnipeg, MB, R3T 5V6, Canada (e-mail:
umghiasn@cc.umanitoba.ca; shaahin.filizadeh@ad.umanitoba.ca).
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TSG.2013.2282773

time series, multiple regression, expert systems, fuzzy logic,

neural network, etc.) [10][15]. These methods mostly use
historical data along with influential parameters on the demand
profile (e.g., climatic conditions, day-type, land usage, etc.).
However, due to lack of actual comprehensive historical data,
efforts for prediction of the charging demand of the upcoming
vehicular load have been mostly done through defining optimal
charging scenarios from the utility and the customers points of
view [16][19]. Generally, realizing temporal and spatial behavior of the charging demand on the power system depends on
such factors as the convenience and cost of charging, adequate
parking time, availability of charging stations and their rating,
and most importantly the charging need of PEVs [20], [21]. It
is also reasonable to assume that the society will not develop
radically different transportation habits based on the charging
requirements of the PEVs. In fact, the PEV technologies and
the utility planners may aim to capture the existing driving
traits of the society and enhance their services in that direction.
Therefore, in the absence of actual measured data of the
charging demand use of real-world driving data in the area
of interest provides a realistic picture of driving habits and
characteristics [21][25]. Layout of roads and traffic patterns
influence not only the driving traits but also vehicular energy
consumption, which together contribute to timing and distribution of charging demand. Local driving data also provides
information about potential parking locations where future
PEVs may be connected to the network for charging.
While PEVs are not widely adopted in a given jurisdiction,
it is reasonable to assume that home plugging will be the
dominant (or in most cases perhaps the only) mode of charging
[26]. As the penetration-level of PEVs increases additional
off-home charging stations should be gradually assigned in
strategic locations in order to fulfill the charging demand of
on-road PEVs. Places where a large number of parking events
occur (e.g., shopping centers) will obviously be more subjected
to vehicular charging, so they can be considered as main candidates for future charging stations and necessary fortifications
of the network.
This paper investigates the potential charging demand in
such high-density off-home parking locations, while considering home charging as the most favorable charging mode.
Using real-world driving data, a location-based fuzzy decision-making algorithm (Sections II and III) is implemented
to predict the probability of charging for a given parking
event. Model sensitivity is then analyzed (Section IV), and
the aggregated potential charging load for two major shopping
centers in the city of Winnipeg [27] is investigated as case
study (Sections V and VI).

1949-3053 2013 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

GHIASNEZHAD OMRAN AND FILIZADEH: LOCATION-BASED FORECASTING OF VEHICULAR CHARGING LOAD ON THE DISTRIBUTION SYSTEM

633

Fig. 1. Flowchart of the location-based vehicular charging load forecasting procedure.

II. LOCATION-BASED VEHICULAR LOAD PREDICTION

Location-based study of vehicular load aims to predict the potential load demand due to PEV charging at off-home parking
locations. Whether or not drivers decide to charge their vehicles
in a specific place depends on various parameters; however, humans decision making process for cases like this often does not
involve precise computations or analyses. Indeed, drivers usually evaluate the situation using their own experience, convenience, and other factors, most importantly economic considerations.
The essence of a fuzzy inference engine is to quantify a complex process, e.g., charging decision making, through reasonable assumptions that capture the experience of the expert (i.e.,
the driver). A fuzzy inference system was proposed in [25] to
simulate a drivers decision-making process for PEV charging.
It used the present state-of-charge (SOC) and parking duration
(PD) as its inputs to predict the probability of charging, independent from the actual place of parking. However, in the location-based prediction method presented in this paper, the actual
driving distance to home (DTH) is also used as an additional
input. This is due to the fact that most drivers would prefer to
charge their vehicles at home rather than at other locations. Several factors contribute to this preference including longer duration of vehicle down-time at home (e.g., overnight), and potentially less expensive electricity. The developed fuzzy inference system takes this preference into account but also includes
provisions for the case when there is doubt whether the current
state of charge of the battery is adequate for the drive to home,
through proper rules.
Fig. 1 illustrates the flowchart of the proposed procedure
for location-based vehicular load prediction. It starts with an
analysis of a real-world driving dataset for the area of interest,
and extracts characteristics of recorded parking events. Then,
a fuzzy decision-making engine uses the extracted statistical
results to predict the average probability of charging for every
hour at the location of interest. This, together with local parking
characteristics and market information, determines the expected
charging demand. The following sections present details of the
procedure.

III. LOCATION-BASED FUZZY DECISION-MAKING UNIT

The three input parameters of SOC, PD, and DTH need to be
expressed with linguistic terms and their corresponding membership functions. Once the inputs are fuzzified, the rules of the
fuzzy system are applied to generate the respective outputs. The
outputs are then combined and defuzzified to yield the output of
the fuzzy inference engine, which is the probability of charging.
The following subsections present details of the membership
functions assigned to each input and output variable [28], [29].
A. State of Charge (SOC)
The SOC is the most readily available input to the driver. It
indicates the amount of stored electrical energy that is presently
available. Fig. 2(a) shows membership functions created for the
SOC. Three linguistic terms, i.e., Low, Medium, and High, are
defined to cover the whole working area of the battery storage.
The on-board vehicle controller only operates the battery within
lower and upper bounds, to ensure its longevity. In the study
presented here a range of [15,85]% is considered (corresponding
to a 70% depth of discharge); that is why the Low and High
membership functions shown in Fig. 2(a) attain a value of 1.0
below 15% and over 85% SOC, respectively.
B. Parking Duration (PD)
Parking duration is a variable that represents the anticipated length of the parking event. Note that most drivers
do not have an exact length of time for parking at the onset
of a parking event. Therefore, it is assumed that they express the parking duration with three linguistic terms (Short,
Average, Long) as shown in Fig. 2(b). For example, the
membership function labeled Average fully encompasses
all parking events ranging
from around 45 minutes to around 3 hours. Therefore, any
parking with a duration in this interval will be equally treated;
this makes the model less sensitive to the deviation between the
actual length of the parking event and its anticipated duration
by the driver.
The linguistic terms for the PD input are highly affected by
the available level of charging. This is because the amount of
charge that can be replenished over the period of charging is

634

IEEE TRANSACTIONS ON SMART GRID, VOL. 5, NO. 2, MARCH 2014

TABLE I
RULES OF THE FUZZY SYSTEM

Fig. 2. Membership functions of the input variables. (a) state-of-charge (SOC);

(b) parking duration (PD); (c) normalized driving distance to home (DTH).

directly determined by ratings of the charger. For example a

parking event whose duration may be considered Short (and
hence not worthwhile) for regular charging may indeed be considered Average if fast-charging facilities exist. The membership functions in Fig. 2(b) are designed with the assumption of
level-1 charging (120-V, 15-A).
C. Actual Driving Distance to Home (DTH)
By experience a prudent driver of a plug-in vehicle may
have a reasonable estimation of how much SOC is required to
drive the vehicle to a convenient charging location, i.e., home
in this study. However, knowledge of the required SOC may
be a higher-than-normal expectation for an average driver and
involves several uncertainties. On the other hand, an estimate
of the driving distance to home is doable for most drivers.
Therefore, in the fuzzy inference system described here the
actual driving distance to home is used as a representative of
the required SOC to drive home.
The DTH assigned to a parking event includes the mileage
of all the subsequent trips during the day before arriving home
for overnight parking. Indeed, the DTH is a factor that enhances
the precision of the charging behavior prediction significantly:
it shows the experience of the driver about daily trips, and their

timing and mileage (i.e., the expert knowledge). It also considers the location of each parking event, thereby making the
decision-making process location-specific.
The DTH input is categorized under three linguistic terms of
Short, Average, and Long [Fig. 2(c)]. Note that the battery capacity plays an important role in the definition of the range for
these linguistic terms. A battery with a higher capacity allows a
longer all-electric range. To avoid development of membership
functions for the DTH for vehicles with different battery capacities, a normalized figure (based on the maximum all-electric
range) for the DTH is used.
In reality the all-electric range of a vehicle will depend on
such factors as the age of the battery, driving patterns of the
driver, and traffic, among other things (see Section III-E for further discussion); thus the actual all-electric range will likely be
less than nominal.
D. Rule Table and Defuzzification
A set of 25 rules for charging decision-making are developed for a Mamdani-type fuzzy model [29] as shown in Table I.
These rules act on the three inputs of the fuzzy system to produce outputs that are then aggregated and defuzzified to yield
the probability of charging for a specific parking event. Before
defuzzification, the probability of charging is described using
seven linguistic terms as shown in Fig. 3. The fuzzy AND is
implemented using the min operator, and the center-of-mass
operation is used for defuzzification.
The design of the rules in this fuzzy system is done with a
view to maintain reasonability of the assumption from a prudent drivers point-of-view. The main consideration in the de-

GHIASNEZHAD OMRAN AND FILIZADEH: LOCATION-BASED FORECASTING OF VEHICULAR CHARGING LOAD ON THE DISTRIBUTION SYSTEM

Fig. 3. Probability of charging (fuzzy system output).

sign of the rule table is that if the SOC of the battery is less than
the drivers estimation of the required SOC to drive the vehicle
home (based on the DTH), the probability of charging increases
with respect to the parking duration.
It should further be noted that when the DTH is Long
both the Medium and Low SOC are treated similarly;
this reflects the fact that a Long DTH most likely represents a parking event in the earlier hours, and from a drivers
point-of-view foreseeing all future trips throughout the rest of
the day might not be possible; therefore an increased chance of
charging is given to the Medium SOC range.
E. Additional Factors
1) Climatic Conditions and Aging: It must be noted that
other factors than the ones considered here may impact a
drivers decision to plug in for charging. Severe climatic
conditions (e.g., extreme heat or cold, humidity, etc.) not only
affect the drivers decision making, but may also affect the
three considered inputs. For example, temperature variations
do impact the chemical reactions within a battery and hence
its SOC, and its total capacity (affecting the DTH), and also
the required heating/cooling energy. Although there are studies
that aim to approximate such variations [30], [31], the level of
detail and complexity required by these methods renders them
infeasible for aggregated long-term vehicular load forecasting,
which also involve a large number of storage units with different properties. Therefore, instead of attempting to augment
the model to directly include such auxiliary effects, the paper
investigates the sensitivity of predicted probability of charging
(output of the fuzzy model) as well as final forecasted load (for
the presented case study) to uncertainties in its input variables
collectively caused by such factors as seasonal temperature
variations, humidity, aging, measurement errors, etc. These
analyses are shown in Sections IV (for the model) and VI (for
the presented case study).
2) Utility Tariffs: The DTH input, which is the preference
factor for the location of charging, accounts for the drivers desire for the least expensive transportation and convenience of
charging. In the design of the fuzzy rule table two cost effectiveness objectives are considered as follows:
a) consumption of electricity has preference over gas because of the lower cost of electricity;
b) home-charging has preference over off-home charging
because of lower cost (as well as convenience);
Although the convenience of charging is not quantitatively
countable, it is possible to include the effect of electricity cost

635

variation by only modifying the DTH input without any further change in the kernel of the developed model. A favorable
charging rate at the time of parking will entice the driver to
charge at an off-home location, and this can be captured by
suitably increasing the DTH input. Conversely a decrease in
the DTH resembles an unfavorable charging rate, and hence a
drivers inclination to charge at home. For example, these can
be due to a utilitys time-variant tariffs for peak and off-peak
hours, and/or incentives offered for off-home charging. However, it should be noted that as long as the two said cost objectives of the model are valid the change in the DTH is expected
to be small; i.e., when the price increases but stays below the
equivalent gas price, or when the price decreases but remains
more expensive than the convenience threshold over which a
driver prefers to charge at home.
The investigation of the effect of price variation is mostly
useful when a more accurate forecast of charging demand is
required for short term, and exact charging costs in each location
and for different time are available.
IV. MODEL PERFORMANCE ASSESSMENT
The developed fuzzy model is designed to capture the drivers
decision making process through a reasonable set of rules acting
on the three inputs. The adopted membership functions and rules
will certainly have an impact on the output of the fuzzy system,
i.e., the predicted probability of charging.
Fig. 4 shows the probability of charging for three representative values of the PD input (30 min, 120 min, and 240 min),
while the SOC and the DTH vary within their respective ranges.
In all three surfaces, the probability of charging increases gradually starting from maximum SOC and minimum DTH. Furthermore the figures show an increasing probability of charging
as the PD increases, as noted by the top left corner of each
figure where the probability of charging increases from 60% for
to essentially 100% for
. These
are reasonable expectations, which are satisfied through the selection and composition of the rules.
In order to assess the performance of the fuzzy model an analysis of the sensitivity of its output is undertaken with respect to
variations of the inputs. This analysis quantifies the expected
deviation of the output when inevitable uncertainties occur in
the estimation of the inputs. These uncertainties may arise due
to factors such as the ones discussed in Section III-E.
Table II presents the results of sensitivity analysis for individual variations of 20% in each input. For example the table
shows that the average change in the probability of charging
(predicted by the fuzzy model) is 5.70% when the SOC input
decreases by 20%.
The table shows that the output is most sensitive to the DTH
input, which encapsulates cost effectiveness and convenience
of charging (home preference factor). The SOC, which captures
the available charge factor, is the second most influential input;
and the least sensitivity belongs to the PD input, which quantifies the worthiness of the available parking time for charging.
The analysis presented in this section shows that the combination of the membership functions and the rules do indeed, and
as intended, lead to make the probability of charging more responsive to the inputs with a higher importance, i.e., the DTH

636

IEEE TRANSACTIONS ON SMART GRID, VOL. 5, NO. 2, MARCH 2014

gender, etc.). Instantaneous latitude, longitude, and speed of the

vehicle in each trip as well as their time and date were recorded.
As indicated earlier, two major shopping centers in Winnipeg
are considered. All recorded parking events in these two locations are extracted from the dataset and their statistical characteristics (e.g., parking time, duration, density, etc.) are derived
and used as attributes of parking events in these shopping centers. The large number of parking events recorded (from different participants) in each of the two locations provides confidence about the validity of the data as an indicator of the entire
population [32].
B. Vehicle Subsystem Models
A backward vehicular model [2] is developed to calculate the
in Joules) used by the vehicle in each
mechanical energy (
trip (defined as the distance travelled between any two consecutive stops). This model is based on Newtons second law of
motion and is shown in (1)(4) below.
(1)
(2)
(3)
and
(4)

Fig. 4. Fuzzy decision making surfaces for three representative PDs. (a)
; (b)
; (c)
.
TABLE II
SENSITIVITY ANALYSIS RESULTS OF THE FUZZY MODEL

and the SOC. It further shows that the robustness of the fuzzy
models output to the variations of the inputs.
V. DATA ATTRIBUTES AND SUBSYSTEM MODELS
In the following two sections, a set of real-world driving data
(for the city of Winnipeg) and the vehicular model used are presented.
A. Recorded Driving Data
Real-world driving data collected from 74 conventional vehicles in Winnipeg [27] is used in the analyses shown in this paper.
This set includes data from participants from different areas of
the city with diverse demographic characteristics (income, age,

are the instantaneous mechanical power

where , , and
(W), speed (m/s), and propulsion force (N), respectively. ,
, and
(all in N) are the aerodynamic drag, rolling
force, and grading resistance, respectively. is the air density
,
is the frontal area of the vehicle
,
is the
is the tailwind speed (m/s),
aerodynamic drag coefficient,
is the road grade, and
is the vehicle mass (kg). In the
simulations presented in Section V, the road grade and the wind
speed are set to zero.
The total required electrical energy from the battery can be
calculated using the consumed mechanical energy, the contribution of regenerative braking, and the efficiencies of different
drive train components, as shown in (5).
(5)
is the total electrical energy (from the battery) and
where
is the total regenerative energy (mechanical) during a trip.
,
,
, and
are efficiencies of the vehicle transmission system, generator, motor, and regenerative braking system
is the amount of energy consumed for
respectively.
heating or cooling (air-conditioning load) the vehicle cabin in
a trip, which will depend, at least partly, on the ambient temperature . Note that other factors such as a drivers choice and
the action of the vehicle controller may impact the amount of
heating/cooling power consumed.
The change in the electrical energy manifests itself as variations of the SOC of the battery. It is, therefore, necessary to

GHIASNEZHAD OMRAN AND FILIZADEH: LOCATION-BASED FORECASTING OF VEHICULAR CHARGING LOAD ON THE DISTRIBUTION SYSTEM

determine the battery SOC given the electrical energy transactions. A simplified expression [33] for calculating the SOC is
given a follows in (6).

637

TABLE III
VEHICLE AND DRIVETRAIN SPECIFICATIONS

(6)
where
is the state of charge at the beginning of the trip,
is the nominal terminal voltage of the battery (V), and
is
the capacity of battery (Ah). To account for the losses that occur
during grid charging (ac-dc converter, plug, etc.) an efficiency
figure of 90% is applied to the drawn power.
VI. SIMULATION RESULTS
The simulation results in this section show the probability of
charging for both weekday and weekend in the two shopping
centers. As an example of the load forecasting procedure shown
in Fig. 1, the expected vehicular load in one of the locations is
shown. The simulation results also include level-1 and level-2
charging scenarios.
A. Simulation Setup and Vehicle Specifications
In the simulations presented it is assumed that vehicles leave
). This is because the
home fully charged (with
typically long overnight downtime of the vehicle is adequate
to fully charge its battery. It also conforms to the underlying
assumption that home is the preferred location for charging.
During the daily trips, the SOC of the battery may decline down
to a minimum of 15%. Although charging may be available
to some PEV owners at other places (such as work place) to
create the worst-case scenario no charging is considered in other
off-home locations prior to arriving in the locations of interest.
Equations (1)(5) show that specifications of a vehicle have
significant effect on its required energy. In relation with the required energy, the battery capacity determines the variations of
the SOC. The battery capacity also directly impacts the all-electric range, which is a determining factor in the DTH input to the
fuzzy system.
In this study three types of plug-in vehicles, namely the
Toyota Prius plug-in hybrid, the Chevrolet Volt, and the Nissan
Leaf, are considered. The Prius and the Volt are representatives
of PHEVs with light-duty and heavy-duty battery storage,
respectively. Nissan Leaf represents EVs with higher capacity
of battery storage (in comparison with PHEVs, which have the
option of switching to gas).
Some adjustments in the developed decision making procedure are required to meet the special conditions of Nissan Leaf
(or other battery electric vehicles, if considered). i.e., it is assumed that if there is doubt whether or not the remaining SOC
is adequate to drive the vehicle home, the driver will have to
charge the battery. This primarily affects conditions when the
SOC is Low or Medium and the DTH is Long.
Table III shows the specifications of the three vehicles considered in the simulation, as well as the efficiencies for drive train
components. Although the efficiency of drive train components
do vary depending to the operating conditions, assumption of
constant efficiency figures is commonly made in high-level vehicular studies [34], [35], and is therefore adopted here as well.
A constant value of 500 W is used to approximately represent the heating/cooling power or additional electrical loads

onboard. Note that an accurate characterization of the actual

heating/cooling power requires data that is not reliably quanis only a small portion of the
tifiable, and also that the
consumed energy during a trip and hence the impact of its variations on the SOC are small. The analysis in Section IV showed
that the probability of charging has only modest sensitivity to
SOC variations, and hence the use of a constant value will not
be detrimental to the validity of the results shown.
B. Weekday/Weekend Probability of Charging and Load
The SOC, PD, and DTH attributes for every parking event
at the selected shopping centers are given to the fuzzy system
to generate a probability of charging. The value of probability
for each parking event contributes to the average probability for
the hour the parking occurs. If parking continues to the next
hour(s), the same probability of charging will be carried to the
next hour(s) as long as the battery is not fully charged.
Figs. 5 and 6 show the predicted average probability of
charging in every hour for the two locations during a typical
weekday and weekend, respectively. The value of probability
in each hour indicates the percentage of the vehicles that are
parked at the given hour and will charge. The zero average
probability for some hours is due to the lack of adequate sample
points in such hours.
As shown in Figs. 5 and 6, the value of probability is mostly
dependent on the type of vehicle (i.e., the capacity of their
battery storage). Probability of charging for light-duty battery
storage (e.g., Prius plug-in) is much higher than the other
two. It is due to that in most cases such vehicles arrive with
nearly depleted battery. The substantially lower probability
of charging for the Volt and Leaf at these off-home locations
is an indication that the daily mileage of these large-capacity
vehicles (prior to arriving in the shopping centers) is likely
to be much less than their all-electric range; this implies that
these vehicles will likely receive the main part of their charging
demand during home plugging.
Apart from the probability for charging, for prediction of the
potential peak load (kW) and energy demand (kWh), the number
of the parking events that occur during the day, their duration,
and distribution among different hours must also be known. The
probability of parking curves for shopping center 1 for both
the weekday and weekend are extracted from the dataset and
is shown in Fig. 7. The procedure for doing so along with the
probability of parking curves for shopping center 2 is given in
[32].
As an example of the prediction process, the potential peak
load is calculated at shopping center 1 for both weekday and

638

IEEE TRANSACTIONS ON SMART GRID, VOL. 5, NO. 2, MARCH 2014

Fig. 7. Probability of parking for shopping center 1 for weekday and weekend.

Fig. 5. Average charging probability for weekday (level 1 charging); (a) shopping center 1, (b) shopping center 2.

Fig. 6. Average charging probability for weekend (level 1 charging); (a) shopping center 1, (b) shopping center 2.

weekend. This example shows how the probability of charging

curves in Figs. 5 and 6 enable calculation of the peak load

for any given number of parking events, types of vehicles,

and market penetration levels of PEVs. In particular, with an
assumption of 20% penetration level of PEVs, and an average
daily number of parking events in shopping center 1 equal to
10 000 it is expected that 2000 PEVs will arrive and park in
this shopping center each day. As seen in Fig. 7, at 4 P.M., the
probability of parking is equal to 9% for a weekday, implying
of the PEVs are expected to be parked
that
at this particular hour. It is further assumed that of the PEVs,
35% are light-duty (similar to Prius), 35% are heavy-duty
(similar to Volt), and 30% are all-electric (similar to Leaf). The
number of PEV types as per the assumed breakdown will therefore be 63 Priuses, 63 Volts and 54 Leafs. The probabilities of
charging at 4 P.M. for each PEV type is then read from Fig. 5(a)
as 47% for the Prius, 20% for the Volt, and 13% for the Leaf.
With a nominal power of 1.8 kW for level-1 charging (with
unity power factor), this results in an expected charging load of
.
Similar calculations can be repeated for other hours.
The expected peak charging power as a function of time for
shopping center 1 for both weekday and weekends are shown
in Figs. 8 and 9 (the traces labeled expected) for the assumed
penetration level and breakdown of PEV types. The procedure
can be easily used to consider other PEVs, other penetration
levels, or other composition of PEV types according to market
acceptance of specific vehicle types in different jurisdictions.
The procedure can also be adopted to analyze the charging load
for each individual day of the week if adequate samples are
available.
Figs. 8 and 9 each also show two additional traces labeled as
lower and upper around the expected peak power. These
curves show the average upper and lower bands of the expected
charging power for a given level of uncertainty in the input parameters. As mentioned in Sections III-E and IV factors such
as temperature variations, aging, measurement error, etc., may
contribute to uncertainty in the inputs to the fuzzy engine and
thereby impact the predicated charging load. In this section simultaneous random variations in the inputs around their nominal values for each parking event are considered and their collective impact on the charging load is evaluated. In particular,
for every parking event a random change of up to 20% is applied to each input with a directional consideration for the SOC
and DTH inputs. In particular two sets of experiments are conducted; the first set considers random changes in the positive

GHIASNEZHAD OMRAN AND FILIZADEH: LOCATION-BASED FORECASTING OF VEHICULAR CHARGING LOAD ON THE DISTRIBUTION SYSTEM

639

Fig. 8. Peak power at shopping center 1 during a weekday (level 1 charging).

Fig. 10. Average probability of charging (weekday, shopping center 1, level-2).

Fig. 9. Peak power at shopping center 1 for weekend (level 1 charging).

Fig. 11. Peak power at shopping center 1 (weekday, level 2).

direction for the SOC and negative direction for the DTH input
(e.g., resulting from a chance of charging before or after the
event), yielding the lower probability of charging. A second
set of experiments is done with negative changes in the SOC and
positive changes in the DTH (e.g., due to severely cold temperatures), leading to the upper probability of charging. The relatively tight placement of the two bands around the expected
trace is an indication of the model robustness to inevitable variations of the inputs.
The predicted vehicular charging power must be added to the
present load profile to obtain the total peak power at the location
of interest. This will be necessary for planning of the distribution network to decide about augmentation of network assets to
prepare for the potential vehicular charging demand.

is due to the fact that the charging decision is mainly dependent to the two other factors (i.e., DTH, and SOC), which remain unchanged. The implications of level-2 charging on the
peak power, however, are significant as seen in Fig. 11, which
shows the peak charging load for the same assumed conditions
of Fig. 8. As seen the expected peak power steeply rises to 620
kW (as opposed to just under 130 kW in level-1 charging). The
power remains significantly higher than level-1 charging for the
entire duration of time. Upper and lower bands similar to
the ones in Fig. 8 are also shown.
Fig. 12 demonstrates the hourly average load demand
curves for the two levels of charging for shopping center 1 and
for the considered composition of PEVs. They are obtained by
considering the duration of each parking event for as long as
charging continues. These curves are indicators of the expected
energy demand for every hour. For example, the curve for
level-2 charging shows an average load of around 400 kW at 1
P.M. This indicates that the charging vehicles between 12 noon
and 1:00 P.M. are expected to receive 400 kWh of energy.
It is observed from Fig. 12 the expected energy demand for
level-2 charging will be much higher. Additionally, it is seen that
its variations are much steeper. This is due to the fact that PEVs
connected to a level-2 charger will draw large amounts of power
over a short period of time and disconnect when fully charged,
which suddenly drops their demand power by a large amount.
This is particularly true for light-capacity PEVs (such as the
Prius) whose battery can be charged from a level-2 charger in
less than half an hour.
Analysis of the parking events in shopping center 1 [32]
shows a large number of short-duration parking events (lasting

C. Probability of Charging and Demand for Fast Charging

To show the impact of fast charging on the potential load
demand, the feasibility of level-2 charging (240 V, up to 30
A) is also considered at shopping center 1. The membership
functions for the PD variable of the fuzzy decision making unit
(Section III-B) need to be modified, as the linguistic terms Short,
Average, and Long are defined based on the charge that can be
received by the battery in a certain time interval. This increases
significantly with a level-2 charger over the same period of time.
All other procedures for calculating the probability of charging
remain unaffected.
Fig. 10 shows the probability of charging using a level-2
charger at shopping center 1. The general shape of these probability curves are not drastically different from those for level-1
shown in Fig. 5(a) despite the effective change in the PD. This

640

IEEE TRANSACTIONS ON SMART GRID, VOL. 5, NO. 2, MARCH 2014

the paper demonstrate the ability of the model in realizing the

assumptions underlying its design.
Utilities can affect the charging demand by proper coordination of i) offering fast chargers, which change the effect of
the PD input on the received charge, and ii) regulating tariffs,
which in effect change the DTH input. The results such as the
ones produced in the paper (Figs. 8, 9, 11, and 12) along with
other economic and technical considerations can be used in the
planning of the scale, location, and type of charging infrastructure, and whether upgrading network assets will be required in
providing such service.
REFERENCES
Fig. 12. Average load demand in shopping center 1 for level-1 and level-2
charging (weekday).

less than 1 hour). While level-1 charging may not provide a

substantial amount of charge for such parking events, level-2
charging is appreciably more beneficial. The large amount of
energy received from a level-2 charger will alleviate part of the
PEVs needs when these vehicles arrive home and are plugged
in overnight. This may in fact assist in reducing the impact of
coincidental charging on the distribution network in residential
areas.
The foregoing analysis of level-2 charging is done with the
assumption that the cost of charging is not affected significantly
by the charging level. In reality level-2 charging is likely
more expensive than level-1 charging. As it was explained in
Section III-E2, it is possible to include the effect of the price
of charging in prediction of the probability of charging by
adjusting the DTH input. Despite this, the given analysis is
still valid because it shows the potential increase of the load
should level-2 charging be made available. It can also be used
in deciding whether level-2 charging is economically viable,
given the revenue that can be obtained from the sale of the
extra energy.
VII. CONCLUSION
A location-based forecasting procedure of vehicular charging
load was presented in the paper. It uses a fuzzy inference system
with three real-world inputs to emulate a drivers decision to
charge at an off-home charging location. The main advantage
of this approach is bringing the drivers experience factor to the
decision making process. The fuzzy system produces an average
probability of charging curve (e.g., Fig. 5), which together with
a probability of parking curve (e.g., Fig. 7) can be used to predict
the vehicular charging load due to any perceived combination,
number, and type of plug-in vehicles. Central to this is availability of data for driving and parking that best characterize the
local patterns. The general procedure outlined in Fig. 1 can be
easily adopted for any other location, any day-type or season, as
long as reasonably reliable driving samples are available. Note
that since plug-in vehicles are still in their early stages of entering the market, there is no large-volume tagged (measured)
data pertinent to the charging behavior of PEV owners. Hence
it is not directly possible to fully validate the proposed models
predictions; however the sensitivity analysis results presented in

[1] C. C. Chan, The state of the art of electric and hybrid vehicles, Proc.
IEEE, vol. 90, no. 2, pp. 247275, Feb. 2002.
[2] M. Ehsani, Y. Gao, and A. Emadi, Modern Electric, Hybrid Electric,
Fuel Cell Vehicles: Fundamentals, Theory, Design. Boca Raton, FL,
USA: CRC, 2004.
[3] W. Su, H. Rahimi-Eichi, W. Zeng, and M. Chow, A survey on the electrification of transportation in a smart grid environment, IEEE Trans.
Ind. Informat., vol. 8, no. 1, pp. 110, Feb. 2012.
[4] W. Kempton and J. Tomic, Vehicle-to-grid power fundamental: Calculating capacity and net revenue, J. Power Sources, vol. 144, no. 1,
pp. 268279, Jun. 2005.
[5] M. D. Galus, M. Zima, and G. Anderson, On integration of plug-in
hybrid electric vehicles into existing power system structures, Energy
Policy, vol. 38, pp. 67366745, Nov. 2010.
[6] S. W. Hadley and A. A. Tsvetkova, Potential impacts of plug-in hybrid
electric vehicles on regional power generation, Electr. J., vol. 22, no.
10, pp. 5668, Dec. 2009.
[7] S. Rahman and G. Sharestha, An investigation into the impact of electric vehicle load on the electric utility system, IEEE Trans. Power
Del., vol. 8, no. 2, pp. 591597, Apr. 1993.
[8] J. Taylor, A. Maitra, M. Alexander, D. Brooks, and M. Duvall, Evaluation of the impact of plug-in electric vehicle loading in distribution
system operations, in Proc. IEEE Power Energy Soc. Gen. Meet., Calgary, AB, Canada, Jul. 2009.
[9] R. C. Green, II, L. Wang, and M. Alam, The impact of plug-in hybrid
electric vehicles on distribution networks: A review and outlook, Renewable Sustain. Energy Rev., vol. 15, pp. 544553, Jan. 2011.
[10] A. A. Sallam and O. P. Malik, Electric Distribution Systems.
Hoboken, NJ, USA: Wiley, 2011, pp. 3365.
[11] H. Hahn, S. Meyer-Nieberg, and S. Pickl, Electric load forecasting
methods: Tools for decision making, Eur. J. Oper. Res., vol. 199, no.
3, pp. 902907, Dec. 2009.
[12] I. Moghram and S. Rahman, Analysis and evaluation of five shortterm load forecasting techniques, IEEE Trans. Power Syst., vol. 4, no.
4, pp. 14841491, Oct. 1989.
[13] D. C. Park, M. A. El-Sharkawi, R. J. Marks, II, L. E. Atlas, and M. J.
Damborg, Electric load forecasting using an artificial neural network,
IEEE Trans. Power Syst., vol. 6, no. 4, pp. 442449, May 1991.
[14] M. Chow, J. Zhu, and H. Tram, Application of fuzzy multi-objective
decision making in spatial load forecasting, IEEE Trans. Power Syst.,
vol. 13, no. 3, pp. 11851190, Aug. 1998.
[15] H. S. Hippert, C. E. Pedreira, and R. C. Souza, Neural networks for
short-term load forecasting: A review and evaluation, IEEE Trans.
Power Syst., vol. 16, no. 1, pp. 4455, Feb. 2001.
[16] P. Fernndez, T. G. S. Romn, R. Cossent, C. M. Domingo, and P.
Fras, Assessment of the impact of plug-in electric vehicles on distribution networks, IEEE Trans. Power Syst., vol. 26, no. 1, pp. 206213,
Feb. 2011.
[17] K. Clement-Nyns, E. Haesen, and J. Driesen, The impact of charging
plug-in hybrid electric vehicles on a residential distribution grid, IEEE
Trans. Power Syst., vol. 25, no. 1, pp. 371380, Feb. 2010.
[18] E. Sortomme, M. M. Hindi, S. D. MacPherson, and S. S. Venkata,
Co-ordinated charging of plug-in hybrid electric vehicles to minimize
distribution system losses, IEEE Trans. Smart Grid, vol. 2, no. 1, pp.
198205, Mar. 2011.
[19] K. Qian, C. Zhou, M. Allan, and Y. Yuan, Modeling of load demand
due to EV battery charging in distribution systems, IEEE Trans.
Power Syst., vol. 26, no. 2, pp. 802810, May 2011.
[20] S. Bae and A. Kwasinski, Spatial and temporal model of electric vehicle charging demand, IEEE Trans. Smart Grid, vol. 3, no. 1, pp.
394403, Mar. 2012.

GHIASNEZHAD OMRAN AND FILIZADEH: LOCATION-BASED FORECASTING OF VEHICULAR CHARGING LOAD ON THE DISTRIBUTION SYSTEM

[21] A. Rautiainen, S. Repo, P. Jrventausta, A. Mutanen, K. Vuorilehto,

and K. Jalkanen, Statistical charging load modeling of PHEVs in electricity distribution networks using national travel survey, IEEE Trans.
Smart Grid, vol. 3, no. 4, pp. 16501659, Dec. 2012.
[22] Z. Darabi and M. Ferdowsi, Aggregated impact of plug-in hybrid electric vehicles on electricity demand profile, IEEE Trans. Sustain. Energy, vol. 2, no. 4, pp. 501508, Oct. 2011.
[23] M. D. Galus, R. A. Waraich, F. Noembrini, K. Steurs, G. Georges, K.
Boulouchos, K. W. Axhausen, and G. Anderson, Integrating power
systems, transport systems and vehicle technology for electric mobility
impact assessment and efficient control, IEEE Trans. Smart Grid, vol.
3, no. 2, pp. 934949, Jun. 2012.
[24] D. Steen, L. A. Tuan, O. Carlson, and L. Bertling, Assessment of
electric vehicle charging scenarios based on demographic data, IEEE
Trans. Smart Grid, vol. 3, no. 3, pp. 14571468, Sep. 2012.
[25] S. Shahidinejad, S. Filizadeh, and E. L. Bibeau, Profile of charging
load on the grid due to plug-in vehicles, IEEE Trans. Smart Grid, vol.
3, no. 1, pp. 135141, Mar. 2012.
[26] ECOtality, The EV Project, Q1-2012 Report.
[27] S. Shahidinejad, E. L. Bibeau, and S. Filizadeh, Winnipeg Driving
Cycle: WPG02, [Online]. Available: http://mspace.lib.umanitoba.ca/
handle/1993/3997
[28] L. A. Zadeh, The concept of a linguistic variable and its application
to approximate reasoning-I, II, III, Inf. Sci., vol. 8;8;9, pp. 199249,
301357; 4380, 1975.
[29] E. H. Mamdani, Application of fuzzy logic to approximate reasoning
using linguistic synthesis, IEEE Trans. Comput., vol. C-26, no. 12,
pp. 11821191, Dec. 1977.
[30] V. Agarwal, K. Uthaichana, R. A. DeCarlo, and L. H. Tsoulalas, Development and validation of a battery model useful for discharging and
charging power control and lifetime estimation, IEEE Trans. Energy
Convers., vol. 25, no. 3, pp. 821835, Sep. 2010.
[31] Y. Xing, W. He, M. Pecht, and K. L. Tsui, State of charge estimation of
lithium-ion batteries using the open-circuit voltage at various ambient
temperatures, Appl. Energy, vol. 113, pp. 106115, Jan. 2014.

641

[32] N. Ghiasnezhad and S. Filizadeh, Characterization of prospective

charging locations of plug-in vehicles using real-world driving data,
in Proc. IEEE Power Energy Soc. Gen. Meet., Vancouver, BC, Canada,
Jul. 2013.
[33] F. U. Syed, M. L. Kuang, J. Czubay, and H. Ying, Derivation and experimental validation of a power-split hybrid electric vehicle model,
IEEE Trans. Veh. Technol., vol. 55, no. 6, pp. 17311747, Nov. 2006.
[34] D. W. Gaoi, C. Mi, and A. Emadi, Modeling and simulation of electric
hybrid vehicles, Proc. IEEE, vol. 95, no. 4, pp. 729745, Apr. 2007.
[35] E. Tara, S. Shahidinejad, S. Filizadeh, and E. Bibeau, Battery storage
sizing in a retrofitted plug-in hybrid electric vehicle, IEEE Trans. Veh.
Technol., vol. 59, no. 6, pp. 27862794, Jul. 2010.

Nima Ghiasnezhad Omran (S12) received his B.Sc. in electrical engineering

from Tabriz University, Tabriz, Iran, and the M.Sc. degree in electrical engineering from K. N. Toosi University of Technology, Tehran, Iran. He joined
the University of Manitoba, Canada, in 2010 where he is presently a doctoral
candidate in the Department of Electrical and Computer Engineering. His areas
of interest include power electronics, power systems planning, distribution network control, and smart grids.
Shaahin Filizadeh (S02M05SM10) received his B.Sc. and M.Sc. degrees
in electrical engineering, from Sharif University of Technology in Tehran, Iran,
in 1996 and 1998, respectively. He obtained his Ph.D. from the University of
Manitoba, Canada, in 2004.
He is currently an Associate Professor with the Department of Electrical and
Computer Engineering of the University of Manitoba. His areas of interest include electromagnetic transient simulation, nonlinear optimization, and power
electronic applications in power systems and vehicle propulsion.
Dr. Filizadeh is a Registered Professional Engineer in the Province of
Manitoba.